US 8150350 B2 Abstract A radio frequency transceiver (
102), including a transmitter (104), a duplexer (108) and a direct-conversion receiver (106) including a mixer (140 and 141). An IIP2 calibration system (170), coupled to the transceiver, includes an IIP2 coefficient estimator (172) for calculating an estimate of second-order distortion intermodulation distortion, and an IIP2 controller (174) for adjusting an IIP2 tuning port of the mixer in the receiver to minimize second-order distortion intermodulation distortion in the receiver that may be caused by the receiver receiving a transmit RF signal leaking through the duplexer.Claims(5) 1. A method of tuning a input-related, second-order intercept point (IIP2) of a mixer in a receiver of a transceiver including a transmitter, comprising the steps of:
determining whether power output of the transmitter is greater than a predetermined power output level;
determining whether a received signal strength indication of the receiver is less than a predetermined received signal strength indication level;
providing values for a gain of an IIP2 estimator and an update rate of a IIP2 controller;
using the gain of the IIP2 estimator, performing an optimization algorithm for estimating an estimated value of an IIP2 coefficient that is proportional to second-order intermodulation distortion occurring in the mixer;
using the update rate of the IIP2 controller and the estimated value of the IIP2 coefficient, performing an optimization algorithm for changing a tuning voltage of the mixer;
waiting a period for a change to occur in one or both of the power output level of the transmitter and a received signal strength indication of the receiver;
after waiting the period, determining whether a change in one or both of the gain of the IIP2 estimator and the update rate of the IIP2 controller has occurred;
based upon one of or both the change of the power output level and the received signal strength indication, determining whether a change in one or both of the gain of the IIP2 estimator and the update rate of the IIP2 controller are required; and
if any such change is required, making a change in one or both of the gain of the IIP2 estimator and the update rate of the IIP2 controller, based upon information stored in a look-up table.
2. The method of
3. The method of
4. The method of
5. The method of
Description This is a divisional of application Ser. No. 12/248,421, filed Oct. 9, 2008. The entire disclosure of prior application Ser. No. 12/248,421 is hereby incorporated by reference. 1. Field of the Invention This invention relates generally to radio frequency receivers, and more specifically to reducing second-order intermodulation distortion in a direct-conversion receiver. 2. Related Art A radio frequency (RF) receiver uses the frequency response of a low noise amplifier (LNA), a surface acoustic wave (SAW) filter and a duplexer to attenuate signals that are far from a center frequency of the receiver sufficiently enough to not corrupt a desired signal. If the LNA and the SAW filter are removed from the analog line-up of the receiver, problems that can detrimentally affect the performance of the receiver may arise. In a transceiver, which comprises a transmitter and a receiver, one such problem is a signal transmitted by the transmitter leaking into the receiver. In a transceiver having only a duplexer to isolate the receiver from the transmitter, there is considerably less attenuation in the receiver of signals at the transmit frequency. A receiver that lacks a SAW filter requires additional and/or tighter constraints on the second-order intercept point (IP2) of the mixer. Without a sufficiently high IP2 of the mixer, the presence of second-order intermodulation distortion (IMD2) substantially reduces the sensitivity of the receiver. The IMD2 results from an unwanted squaring of the transmitted signal at the mixer of the receiver. Most cellular wireless transceivers use a direct-conversion receiver because a high level of integration can be obtained. However, a direct-conversion receiver requires a high input-related second-order intercept point (IIP2), which is the theoretical input level at which the power of the IMD2 products are equal in power to the power of a desired signal. When the receiver is at sensitivity and the transmitter is at maximum output power, the self-blocking effect of the transmitter can, through a second-order nonlinearity of the mixer, desensitize the receiver. In such a receiver, transmitted signals are attenuated via the duplexer by approximately 50 dB; nevertheless, attenuated transmitted signals leak into the receive signal path prior to a front-end amplifier. For example, in the receiver, the duplexer attenuates a strong signal, transmitted by the transmitter of the transceiver, of +25 dBm (316 milliwatts) located at 190 MHz from the center frequency by only 50 dB, thus resulting in a signal of −25 dBm (3.16 μwatts) at the input of the front-end amplifier. This −25 dBm signal creates strong IMD2 products that land on the desired signal, thus producing interference. The IP2 of the mixer in a direct-conversion receiver that lacks a SAW can vary due to manufacturing processes and/or change in temperature. Any mismatch between differential signals in the mixer causes a reduction in the IP2 from an optimal IP2. The mismatch can be due to variations in manufacturing processes or due to temperature changes during operation of the receiver, or due to both causes. The mismatch can also occur due to direct current (DC) offset, local oscillator leakage, or other factors. When there is a large mismatch between differential signals in the mixer, a worst case IP2 of approximately 25 dBm can occur. The following example uses the worst case IP2 of 25 dBm from measured data of a known 3G receiver. IMD2=P When a transmit signal is at maximum power, 25 dBm, the IMD2 referred to the input of a transconductance amplifier (not shown) is −75 dBm. The received power spectral density, Ior, as per the sensitivity specification set forth in the 3rd Generation Partnership Project (3GPP) standard, should be at or below −106.7 dBm to achieve a 0.1% bit error rate. The thermal noise, kTBF, in this example is approximately −99 dBm. Because the power of the IMD2 over the bandwidth of the desired signal is much greater than the kTBF, the sensitivity rises to −82.7 dB, i.e., 24 dB above the required sensitivity. Therefore, a receiver, including, in particular, a direct-conversion receiver that lacks a SAW, should have a sufficiently large IP2 to meet the specifications of the 3GPP standard. The present invention is illustrated by way of example and is not limited by the accompanying figures, in which like references indicate similar elements. Elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. The receiver In addition to a desired component downconverted to baseband from a desired received RF signal, the signal outputted by the I-channel mixer In general, the output z(t) of a nonlinear device can be represented as follows:
where s(t) is the signal at the input of the nonlinear device, and a The output y The actual I-channel mixer It can be shown also that the second-order intermodulation distortion product for the I-channel mixer is
The signal x(t) that is transmitted by the transmitter Alternatively, it can be represented as:
At the I-channel mixer After substituting equation (5) into equation (7), yields
Performing the multiplication indicated in equation (8), yields the following equation (9):
If there were no second-order intermodulation distortion products produced in the I-channel mixer However, second-order intermodulation distortion products may be produced in the I-channel mixer Substituting equation (9) into equation (11) yields the following equation (12),
Performing the indicated squaring operation in the above equation, and removing all cross terms because they result in higher frequency terms that will be filtered out by the analog filter
Removing the double-angle terms after squaring, and saving only the terms located at or near DC, from equation (13) yields,
In a similar fashion, it can be shown that
Therefore, the total in-band second-order distortion product, y
At the IIP2 intersection, s(t)=V If the IIP2 is measured in V
If the IIP2 were measured in decibels V
In a similar fashion, it can be shown that a
Therefore, minimizing the value of a The output from each lowpass filter The baseband signal in the receiver The calibration system The estimator The estimator The update equations In one embodiment, a least mean squares (LMS) based optimization algorithm is used in the estimator The transmit-to-receive gain of the transceiver The selection of a step size of the controller The estimator In one embodiment, the DC estimator Depending on the embodiment, the preceding I The receiver The DC estimator A firmware-based supervisor algorithm (see The calibration algorithm of the estimator The controller The LMS algorithm to minimize a Upon substituting an estimate of â
A first order numerical approximation of the LMS gradient can be expressed as follows, with k being the latency between v
The sign, alone, of the gradient can be used in the controller
Equation (32) can be substituted into equation (30). Equation (30) then becomes:
Alternatively, the gained estimate of a This replacement allows the algorithm to be desensitized from variations in the linear gain, G Alternatively,
Equation (34a) can be substituted into equation (33). Equation (33) then becomes:
where k represents a delay, measured in number of samples, for a reaction to occur in â The controller The I-channel controller The output of the sign element The output v However, for the sign element Equation (36) and equation (38) are multiplied at the gradient direction element The output of the gradient direction element If {−1, −1} or {1, 1} then the gradient direction is −1. Else if {+1, −1} or {−1, +1} then the gradient direction is +1. Otherwise, the gradient direction is 0, i.e., sgn{Δâ The output of the gradient direction element Maximum SNR means that the transmitter P As the transmit power reduces, the performance of the calibration system As received signal strength increases above sensitivity, the performance of the calibration system The desired receive signal and in-band blockers act as uncorrelated “noise” sources. The larger the power of the desired receive signal plus the in-band blockers is, the smaller is the SNR recognized by the estimation algorithm. The “linear” residual baseband component dominates SNR recognized by the estimation algorithm. As the signal strength of the desired signal plus any in-band blocker increases, the gain μ As the transmit-receive isolation of the duplexer The calibration system The calibration system The calibration system In one embodiment, the transceiver Although the invention is described herein with reference to specific embodiments, various modifications and changes can be made without departing from the scope of the present invention as set forth in the claims below. For instance, although the exemplary embodiments show that the invention is used with a direct-conversion 3G receiver Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of the present invention. Any benefits, advantages, or solutions to problems that are described herein with regard to specific embodiments are not intended to be construed as a critical, required, or essential feature or element of any or all the claims. Unless stated otherwise, terms such as “first” and “second” are used to arbitrarily distinguish between the elements such terms describe. Thus, these terms are not necessarily intended to indicate temporal or other prioritization of such elements. Patent Citations
Non-Patent Citations
Referenced by
Classifications
Legal Events
Rotate |